[Princess Garnet]

For those who remember, I "sold" my secondary PC to a friend a good while back. Well, he didn't have it fully paid off (I told him I'd wait originally), and he recently lost his job (again), so he decided to give it back to me and have me give him and old basic one.

Anyway, I put my E8400 in it, and while I remember it having temperature issues (not necesarilly high temperatures) in when it was in my primary PC (with the same heatsink it has on it now), this seems a bit odd to me.

Maybe I'm just too used to seeing my E8600 and both of it's cores always being at the same exact value (plus or minus two at the most) to think this is acceptable? Or is this really something to be worried about?

Note that it's at stock speed (less actually since Intel SpeedsStep is currently in effect during that screenshot), and it's only at it's listed value of 1.2V (1.2250V set in BIOS). The CPU itself is fine, but the first core is a good 8-10 degrees warmer, and the second is a good 9-11 degrees warmer than the first core. This means there can be a 19-21C difference between the CPU itself and the second core.

I know that the cores themselves usually run warmer, and that one can often be warmer than the other, but I do not ever remember seeing it like this when I used it before (and again, it has the same heatsink now as it did then). The CPU is even lapped, and I've tried fitting the heatsink (Artic Cooling Freezer 7 Pro) twice, and that didn't change it. This is also at idle too. Under load, I imagine a good 10 to 15 degrees will be added, putting one core in questionable range.

 

[Squall-Leonhart]

The thermal grease has not applied evenly across the spreader.

 

[Princess Garnet]

Well, I tried applying twice, once with a small dot in the center, and another time spreading it thinly across the whole CPU. Both methods gave me about the same temperatures. I know it hasn't had time to cure yet, but unless that's going to drop it a good deal (it usually only drops it ~3C I think), then letting it cure isn'tthe problem.

 

[Squall-Leonhart]

have you checked to find out if something is using the second core at all?

 

[Princess Garnet]

Nothing is from what I see (it's basically a fresh install of Windows 7 RC 32-bit), and right now, the temperatures are 29C for the CPU, 37C for the first core, and 48C for the second core. I am using Windows 7 on it versus it being checked under Windows XP when I used it for my primary PC in the past, but I'm doubting that's the sole cause for this.

The PC itself is clean (I cleaned it when I got it back from him, although it wasn't that bad), before anyone mentions to check that.

 

[Squall-Leonhart]

check the latency process with proces explorer.

 

[Cid Highwind]

One of the cores of my Quad is always a good 8 degrees warmer. For a Quad it seemed to be common, I don't know about yours though. Seeing how my Quad consists of 2 dual cores, it wouldn't be too alarming as long as it doesn't go over the limit and cause instability.

 

[Squall-Leonhart]

Thats wierd :\ theres only a 3c variance from my coolest to my warmest.

 

[Princess Garnet]

Now how much warmer is that warm core than the "CPU" value itself? Is it idling not far under 50C, and is ~20C higher than the "CPU" value, and with a decent cooler?

I dug through my Photobucket account and couldn't find much, since I deleted alot of old stuff from it recently to clean it up, but I did find one that I'm pretty sure were temperatures of my E8400 and not my current E8600, since the cores of it were never much higher than the CPU itself (if the core temperatures go up, so does the CPU). I still had my two 320GB drives too, and I'm sure I got my third and fourth 640GB drives before my E8600, but, I could be wrong. I miss those 39C Northbridge temperatures too.

This is my E8600.

Seriously, the cores are cooler than the CPU value itself, and the CPU is the coolest thing in my PC (well, tied with hard drives). Talk about backwards.

That first one shows what I recall to be about what I saw with the E8400 when I primarily used it. I know it was sometimes normal for one core to be 2C-3C warmer than the other, but I knew this stair step effect going up in 10s between each was out of it. If it were to go under load, one core will be around the 60C range I'd guess, plus or minus some, and I think 62.xC is the rated maximum, right? I don't like that for a stock frequency CPU with a decent cooler.

Squall-Leonhart, I'll do that when I have it hooked back up. I'm downloading the expansion pack for Sims 3 from Direct2Drive right now on my primary PC. Chances are, I'll try reseating it one more time. Would you recommend applying the paste with the drop method, or spreading it thinly before putting the cooler on. I usually get close results either way.

 

[Squall-Leonhart]

Multicores are better with the line method, just determine which direction the cores lay in first then mark the line across the both of them.

 

[xRyudo]

Hmmm just did a check with mine.

Intel Pentium Dual-Core E5200 is my CPU.

Not much difference in temperatures there. It's mostly like Squall said.

 

[ElijahTW]

Maybe the headspreader isn't completely even.

 

[Squall-Leonhart]

yeah, and thats a ***** to check, you need thermal adhesive to even reapply it.

 

[Princess Garnet]

You know, now that I think more about it, this was a very, very early E8400, and I do remember the temperatures being off. I remember testing it up to ~70C, and then tried touching the base of the heatpipes on the cooler, and it wasn't even warm at all, so I'm not sure if the temperatures are right or not. I remember it started as an E8400 on an Artic Cooling Freezer 7 pro, and upgrading a Xigmatek HDT s1283 did not help, and lapping the IHS did not help. I'm going to go with three things, or any combination of them.

1. The IHS isn't on the CPU too well.

2. The thermal sensors are wrong.

3. It just runs warm period no matter what.

Multicores are better with the line method, just determine which direction the cores lay in first then mark the line across the both of them.

I can't believe I completely forgot about that, but given the above, I'm not sure if it's worth bothering with it. I have a big feeling it won't help, but I may try again anyway.

Hmmm just did a check with mine.

Intel Pentium Dual-Core E5200 is my CPU.

Not much difference in temperatures there. It's mostly like Squall said.

Your results are abnormal compared to what most people get, and more like mine on my E8600. Usually, the cores run at least a good 3C-5C warmer (sometimes more) than the CPU itself, and often one core runs a bit warmer than the other core (in my experience, it's core #1, or the second core, that runs warmer for most people). Having cores running cooler than the CPU itself, and with such little variance between all of them, seems to be against the norm, but in a good way, of course.

 

[tuanming]

I wouldn't worry too much about it as long as it work fine.